3.1. Materials and Methods
Twelve individuals of four tree species were potted in May 2019 in 7-liter pots with well-drained soil in the greenhouse in Tulln, Austria. Two weeks before enclosing the trees in the gas chambers the trees were moved to a nearby greenhouse in Vienna. Drought and salt treatments were started 10 days before the gas measurements. Drought stressed trees (d) were kept at 5% volumetric water content, whereas the control (c) was held well-watered at 100% field capacity with 13.4% volumetric water content. Salt stressed trees (s) were infused with 5L of 50mM NaCl two and one week before the gas measurement and re-watered every second day. 24h before enclosing the trees into the gas chambers, they were rinsed wioth water to remove dust and finger-prints and moved to a climate chamber (Fitotron, Weiss Gallenkamp, UK) at 25°C, 60% rH and 300µmol/m²s PAR (photo…etc.) to adapt to the air temperature inside the gas chamber.
3.2. Data collection
The measuring campaign took place in June and July 2019. We measured O3 and CO2/H2O gas fluxes, SWP and Na and Cl leaf content. The trees were enclosed in gas exchange cuvettes of 10 L capacity. Each tree was pre-fumigated with [O3] between 75 -170 ppb for 30 min before measurement. Three of the cuvettes were used for plant measurement and the one left was used as a reference, remained empty. Each day six trees were measured, three in the morning and three in the afternoon, one tree per chamber. For avoiding selection bias, the tree species were randomized. After being fumigated and monitored in the chambers, the tree leaves were cut off and their fresh weight, leaf area, dry weight, and chloride content, were examined individually. The following variables were measured:
• Leaf Temperature – LT (° C)was measured using a thermocouple directly on the underside of a leaf located in the middle of the tree during the gas exchange measurements.
• Transpiration rate - E (mmol/m²s): Using the CIRAS-3 SC CO2/H2O Gas Analyzer, it was calculated from the next equation:
27 𝐸 = [𝑊 × (𝑒𝑜𝑢𝑡− 𝑒𝑖𝑛)
(𝑃 − 𝑒𝑜𝑢𝑡) ] (5)
Where W is the mass flow of air entering the cuvette per unit leaf area, ein is the partial pressure of water vapor of reference air supplied to the cuvette, eout is the partial pressure of water vapor in the air inside the cuvette, and P is atmospheric pressure (PP Sytems, 2018).
• Stomatal Conductance – gs (mmol/m²s): Using the CIRAS-3 SC CO2/H2O Gas Analyzer, it was calculated from the next equation:
𝑔𝑠= 1
𝑟𝑠× 103(𝑚𝑚𝑜𝑙
𝑚𝑜𝑙 ) (6)
Where rs represents stomatal resistance and is explained in the next equation, eleaf is saturated water vapor pressure inside the leaf at a reference temperature, eout is the partial pressure of water vapor in the air inside the cuvette, E is transpiration rate, P is atmospheric pressure, and rb
is the boundary layer resistance to water vapor (PP Sytems, 2018).
𝑟𝑠 (𝑚2 𝑠 𝑚𝑜𝑙 −1) = [ (𝑒𝑙𝑒𝑎𝑓− 𝑒𝑜𝑢𝑡)
(𝐸 × (𝑃 − (𝑒𝑙𝑒𝑎𝑓 + 𝑒𝑜𝑢𝑡)/2))] − 𝑟𝑏 (7)
• Ozone conductance - gO3(mmol/m2s): It was calculated according to Fares et al (2008) where ozone conductance represents the deposition of ozone in the plant (R1), the ozone loss on the walls of the chamber (R2), and the product of the gas-phase reactions of ozone with BVOC (R3).
These last two depositions were removed from the calculation as R2 was not considered in the experiment: ozone uptake was obtained from the difference of outgoing ozone from a chamber with a tree inside and the outgoing ozone from a chamber without plant inside. As well, R3 was not considered, as there was no measurement of BVOC in this study. Additionally, as found by Fares et al (2008), gas-phase reaction would only be significant for F. sylvatica and B. pendula, as they are mono- and sesquiterpene emitters.
• Net Photosynthesis – A (µmol/m²s): Using the CIRAS-3 SC CO2/H2O Gas Analyzer, it was calculated from the next equation:
𝐴 = −[((𝐶𝑜𝑢𝑡− 𝐶𝑖𝑛) × 𝑊) + (𝐶𝑜𝑢𝑡× 𝐸)] (9)
28
Where Cout represents CO2 concentration exiting the cuvette, Con represents CO2 concentration entering the cuvette, W is the mass flow of air entering the cuvette per unit leaf area, and E is transpiration (PP Sytems, 2018).
• Stem water potential – SWP (MPa): After gas exchange measurements, one leaf was selected from the tree, was covered with aluminum foil and left in a plastic bag for 30 minutes. Later SWP was measured using a pressure bomb as explained in Williams and Araujo (2002).
• Leaf area – LA (cm²): After gas exchange measurements were done, all leaves of the tree were cut, excluding the brown ones, and scanned. Posterior, the images were analyzed using the software WinFOLIA 2014a.
• Ozone uptake - ΔO3 (µg/m³): Using the Ozone Monitor BMT 932 it was calculated from the next equation:
𝛥𝑂3= [𝑂𝑖𝑛− 𝑂𝑜𝑢𝑡] (10)
Where Oin represents O3 concentration entering the cuvette and Oout represents O3 concentration exiting the cuvette (BMT MESSTECHNIK GMBH, 2014).
• Ozone uptake per leaf area - ΔO3 (µg/m³cm²): As appreciated in equation 9 but corrected with the leaf area of each tree (LA):
𝛥𝑂3= [𝑂𝑖𝑛− 𝑂𝑜𝑢𝑡
𝐿𝐴 ] (11)
• Ozone losses - ΔO3 (%): Taken from Fares et al. (2007) where ozone is displayed as the percentage of uptake:
𝛥𝑂3= [𝑂𝑖𝑛− 𝑂𝑜𝑢𝑡
𝑂𝑖𝑛 ] (12)
• Chloride Content – Cl (mg/gDW): After gas exchange and leaf area measurements were done, 0,5 g of fresh weight leaves where shredded using liquid nitrogen in a hand mortar. Later, the powder was added to a 25 ml distilled water and left for 1 h on a shaker. Subsequently, it was filtrated through a filter paper following a final analysis in an ion chromatograph, 881 Compact IC pro – Anion, and this supplement: Autosampler: 858 Professional Sample Processor.
3.2.1. Gas exchange cuvette
The chamber was a 10 L oven bag which can maintain the gas inside an was easy to manipulate. For avoiding low irradiance, the lamps that were located on the top of the chambers were covered with an aluminum layer of 50 lengths, to improve reflection also on the sides of the plants. Furthermore, the system was constructed with tubes and connectors made of inert material, like Polytetrafluoroethylene
29
(PTFE), and all the surfaces that could react with O3 or VOC within the track in which the ozonated air was flowing were covered with Teflon tape. All measurements were performed at 1000 µmol/m² s.
As explained in section 3.2, one chamber was not used for tree measurements but as an ozone reference, and before connected to the BMT, a 1L/min valve was previously installed. Additionally, as four outlets were connected to the CIRAS a relay module was used with four mechanic valves in which every 20 minutes one of the valves was opened and the rest closed. See Figure 7.
Firstly, an 80 L/min pump vacuumed air from the exterior of the building. This air was dehumidified at 5°C and later cleaned in a charcoal filter. After, the air was divided into two channels, one of them with a 1L/min flow valve connected to the ozone generator (which function properly with a maximal flow of (L/min) and the other one linked afterward again to this fumigated air and separated into five channels, one directed to the CIRAS-3 SC CO2/H2O Gas Analyzer as a reference with a valve of 1L/min previously installed, and the four remaining to the four chambers, also with valve previously installed (with a max.
flow of 10L/min but a real one of 7.88L/min, see below), each of them with four outlets: one for overflow, one for temperature, one for ozone measurements directed to the Ozone Monitor BMT 932 and the other aimed to the CIRAS for CO2 and H2O measurements.
Each chamber counted with a flow of 7,88 L/min flow and the safety O3 channel with 1,65 L/min. The chamber flushing time was calculated using the following equation from Niinemets et al. (2011), where a time of 4τc is needed for 94% of full system response, in this case: 3 minutes and 30 seconds.
𝜏𝑐(𝑚𝑖𝑛𝑢𝑡𝑒𝑠) = ln 2
𝐹/𝑉 (13)
Where τc is the chamber flushing half-time, F is the flow rate through the system and V is the chamber volume. This equation is based on ideal turbulent mixing in the chamber without the plant, nevertheless, as the volume of each chamber was 10 L, a relatively small one in comparison with other studies, the steady conditions should be reached on this time (Calfapietra et al., 2016).
30 Figure 6. Enclosed system for gas exchange measurements
3.3. Data analysis
In each specie, an Analysis of Variance (ANOVA) followed by Tukey’s test was performed contrasting drought and salinity treatments against control treatment. Additionally, the analysis of correlation was performed between ΔO3 and gs, gO3, and SWP individually for all species. The analysis was done in R, for linear, logarithmic and exponential relations, and choosing the one with a significance p_value < 0,05 and the bigger R2.